Why a hydrogen economy is hard to start: what the Air Products cancellation actually shows

Air Products just wrote off up to $2.9 billion on a hydrogen project it had been building in Louisiana, and the stock went up nine percent on the news. That reaction is the interesting part. To understand why the market clapped, you have to understand what the project was trying to do, and why the thing it was trying to do is genuinely difficult, in ways that have very little to do with chemistry.

I want to walk through the mechanism. Not the deal, not the writedown — the underlying problem the project was trying to solve, and why it stalled.

The project, in one paragraph. The Louisiana Clean Energy Complex (LCEC) was going to make hydrogen at industrial scale, 750 tonnes a day, from natural gas, capturing the CO₂ produced along the way and burying it underground. That is called blue hydrogen: hydrogen made from fossil fuel, but with the carbon caught rather than released. The plan was to sell that hydrogen mostly to fleet operators, trucks, buses, industrial vehicles, as a diesel replacement. On June 30, Air Products announced it was walking away, citing "slower-than-expected development" of hydrogen mobility markets.¹ That phrase is doing a lot of work. Let me unpack it.

The bootstrapping problem

Hydrogen is not like petrol. When you buy a diesel truck today, you are stepping into an infrastructure that has been quietly compounding for a century: refineries, pipelines, tank farms, filling stations on every arterial road, mechanics who know the engine, insurers who know the risk profile. None of that exists for hydrogen at anything like the same density.

To run a hydrogen fleet, you need three things to exist at once:

1. Vehicles that burn hydrogen (fuel-cell trucks, buses, forklifts).
2. Refuelling stations near the routes those vehicles run.
3. Production and delivery at a price that competes with diesel.

Each of these depends on the other two. Fleet operators will not buy fuel-cell trucks if there is nowhere to refuel them. Station operators will not build stations if there are no trucks to serve. And producers like Air Products will not build a $4.5 billion plant unless there is confident demand from stations and fleets to soak up 750 tonnes a day.

Petrol had the early automobile industry as the shove. Natural gas had utilities and municipal gas networks. Electricity had, quite literally, Edison building the whole stack, generation, distribution, and light bulbs, because no one piece worked without the others.

Hydrogen mobility has not yet found its shove. The 45V clean-hydrogen tax credit in the US was supposed to be part of it, but the implementation rules stayed contested into 2026, and light-duty fuel-cell vehicles never took off. Heavy trucks and buses are still moving, but slowly, and not at the volumes a 750-tonne-per-day plant needs.

That is what "slower-than-expected development" means in the press release. It means Air Products built the production side of a three-legged stool and the other two legs did not arrive on schedule.

Why blue hydrogen, and why it matters here

A quick definition, because this is where the piece gets specific.

Green hydrogen is made by running electricity through water and splitting it into hydrogen and oxygen. If the electricity comes from renewables, there is no carbon in the process. It is expensive because electrolysers are expensive and renewable electricity, at the scale you would need, is not yet cheap enough.

Blue hydrogen is made the old-fashioned way, reacting natural gas with steam to strip out the hydrogen, but with a carbon-capture unit bolted on to trap the CO₂ before it reaches the atmosphere, and a sequestration site to store it underground. It is cheaper than green hydrogen today, but it carries two liabilities: it depends on the price of natural gas, and it depends on carbon capture actually working at commercial scale, which is a whole other bootstrapping problem.

LCEC was blue. Air Products was betting that blue hydrogen could be the transitional product — cheap enough to build a market, clean enough to qualify for subsidies, familiar enough that industrial customers would sign long-term offtake agreements. That bet did not fully close. The customers did not sign at the volumes required, and the regulatory picture for what counts as "clean" hydrogen stayed muddy long enough that project economics could not be locked down.

The metaphor I would reach for: Air Products was trying to build a bridge from fossil hydrogen to green hydrogen, and the far bank kept moving.

Why the market cheered

Now we can look at the 9% stock pop.

Writing down $2.9 billion is not, in itself, good news. It is an admission that money already spent, on engineering, permitting, contracts, early construction, will not produce a return. In accounting terms, it is an impairment charge: a one-time recognition that an asset on the balance sheet is worth less than the company had been carrying it at.

The reason the market rewarded it is separate from the loss itself. Investors had been watching Air Products commit to megaprojects for years under its previous CEO, and had been increasingly nervous that hydrogen demand would not show up in time to service the debt those projects implied. New CEO Eduardo Menezes, in office since late 2024, has now killed the biggest of them.²

The 9% pop is essentially a vote on capital discipline. Investors were saying: we would rather see you take a large one-time hit and stop bleeding capital into a market that has not developed, than watch you keep spending in the hope that it will.

That is a hard signal about how the market currently views large-scale clean-hydrogen production. Not that the chemistry is wrong. That the demand-side coordination has not happened, and the wait is not worth financing.

The pivot that came with it

Announced in the same breath as the cancellation: Air Products is finalising a deal with Yara and NEOM to distribute renewable ammonia from Saudi Arabia.¹

Ammonia (NH₃) is turning out to be a useful workaround for hydrogen's transport problem. Pure hydrogen is a nightmare to ship — it is the smallest molecule in the universe, it leaks, and it needs to be either compressed to enormous pressures or cooled to near absolute zero to move usefully. Ammonia, by contrast, is a familiar industrial liquid with an existing global shipping fleet and port infrastructure. You can make ammonia from green hydrogen plus nitrogen, ship it, and either use it directly (as fertiliser or fuel) or crack it back to hydrogen at the destination.

By pivoting to distribution, Air Products moves from being the builder of expensive production plants to being the buyer of somebody else's low-carbon molecules. It keeps the company in the low-carbon business without carrying the upstream capital risk. Whether that turns out to be the right position depends on whether the demand side ever shows up. If it does, distributors do well. If it does not, at least the writedown is behind them.

What to watch

Two things.

First, whether the heavy-truck and maritime hydrogen markets pick up enough over the next two to three years to justify anyone building another LCEC-scale plant. If they do, this cancellation will look like bad timing. If they do not, it will look like the moment the industry admitted the light-duty hydrogen story was over.

Second, whether ammonia-as-carrier becomes the actual shape of the hydrogen economy, rather than pure-hydrogen pipelines and refuelling networks. The Yara/NEOM deal is a small signal in that direction. Others will follow, or they will not, and that will tell you a lot about which version of the transition is happening.

Footnotes and links

Further reading

• Transport Topics, "Air Products scraps Louisiana hydrogen project": https://www.ttnews.com/articles/air-products-hydrogen-louisiana
• Investopedia, market coverage of the announcement, 30 June 2026: https://www.investopedia.com/stock-market-today-dow-jones-s-and-p-500-06302026-12009393